125 Język. Komunikacja. Informacja Language. Communication. Information I. Koutny, P. Nowak (red./eds.) 8/2013: 125 132 Ida St r i a Classifications of artificial languages Abstrakt (Klasyfikacje języków sztucznych). Istnieje wiele systemów nazywanych językami sztucznymi. Niektórzy autorzy wymieniają nawet ok. 1000 (por. Blanke 1989, Schubert 1989), a ich liczba nadal rośnie. Ta liczba może się jednak diametralnie zmieniać w zależności od definicji, jako że najszersza obejmuje systemy od kodów ograniczonych aż po szczególne formy języków naturalnych. Ninejsza praca próbuje odpowiedzieć na pytanie, czym tak naprawdę jest język sztuczny oraz sklasyfikować je według różnych kryteriów. Proponuje się także rozmieszczenie ich na skali sztuczności. Abstract. There are many systems called artificial languages. Some authors mention around 1000 of them (cf. Blanke 1989, Schubert 1989) and the number is still rising. However, this number might be changing radically depending on the definition, as the broadest comprises systems from restricted codes to particular forms of natural languages. The present paper tries to answer the question of what an artificial language really is as well as classify them according to different criteria. It is also proposed that those languages should be arranged on a scale of artificiality. The title of this paper needs clarification. What exactly is meant by artificial languages? What is the difference between artificial and non-artificial? Already this distinction suggests a binary opposition, which, as will be later seen, is a great simplification of the classification of languages. As a matter of fact, this dichotomy of natural i.e. traditional ethnic languages such as English or Hindi and artificial languages is relatively new as it dates back to 19th-century linguistics which was strongly influenced by Darwin. However, when language is considered as a construct of human culture, the label natural does not really seem to apply. In addition, the term artificial is often misused by being applied to a wide variety of codes and systems not related in any way. Detlev Blanke lists several different meanings: 1. Regularized and standardized literary language, as distinguished from dialects [ ].
126 Ida Stria 2. Ethnic languages, highly regularized to maintain them at a particular stage of development (Sanskrit, Church Latin) or to modernize them (Modern Hebrew, Bahasa Indonesia, Landsmål). 3. Consciously created languages to facilitate international communication [ ], that is, planned languages. 4. Nonredundant, formulaic, or symbolic languages to facilitate scientific thought [ ]. 5. Programming languages for computers [ ]. 6. Machine languages for automatic translation. (Blanke 1997: 3). The types 4 6 will not be a part of this paper as they are considered here to be highly restricted subsets of a language in line with Schubert (1989: 9). 1 In the interlinguistic literature the term artificial as opposed to natural is regarded as crudely misleading (Schubert 1997) because it suggests that languages created to facilitate international communication are in fact identical to machine or formulaic languages. Other names have also been used throughout history: universal, international, auxiliary, constructed, planned, and invented. Blanke (1989) distinguishes between two groups of terms that sometimes overlap, that is constructed/artificial which points towards the creation of the language and planned/universal which describes the language s function. Under the name artificial a good deal of various items can be placed (cf. Eco 1997: 2n., Carlevaro 1989: 177): programming or machine languages (COBOL) experimental languages: philosophical (Toki Pona) logical (Lojban) pasygraphies (Solresol) international auxiliary languages (planned languages, hereafter IALs) artistic languages (Klingon, Quenya) normative languages: superdialectal (Rumantsch Grishun, Standard Arabic) standard literary languages revived (Cornish) controlled languages (Caterpillar Fundamental English) reconstructions (Proto-Indo-European) pidgins and creoles oneiric languages, xenoglossias, glossolalias. This paper uses the term artificial as a broad name for all languages that are the result of deliberate and conscious creation and planning. Out of the constructs listed, the first type will not be considered for reasons already mentioned and the last type will not be considered because it does not meet the requirement of being deliberately created. 1 Types 1-2 are often excluded from classifications because of being only regularized versions of original ethnic languages.
Classifications of artificial languages 127 1. Some traditional classifications In his article Blanke (2001: 51n., see also Blanke 1985: 99-110) gives six different types of classifications of artificial languages according to: realization level (graphic/phonetic, i.e. pasygraphies or universal scripts vs. pasylalies or universal languages), material and structure, i.e. invention vs. imitation linguistic concept, i.e. how well the project is formulated linguistically level of details/development availability role in real communication. While the first two are common enough, the next four do not appear in the literature. However, the most widespread and quoted classification is the second (regarded as traditional), proposed by L. Couturat and L. Leau in 1903/1907 (cf. Schubert 2011: 50). It is a very crude division into three main types: a priori (based on invented elements) a posteriori (based on elements already given) mixed This proposal has been worked on by several linguists. A more detailed version can be found in Janton (1993: 6n.), where the author presents the following categories: a priori, (i.e. philosophical languages) a posteriori simplified ethnic (i.e. minimal) mixed schematically derived (Volapük) with partly schematic and partly natural derivation (Esperanto) naturalistic with some schematic traits (Novial) with natural derivation (Occidental, Interlingua) It can be seen that these categories mirror a scale of artificiality. The poles determine whether a project is derived from (an) ethnic language(s) or deliberately constructed. It is immediately noticeable that the examples given include only the universal and international auxiliary languages (that is to say, planned languages according to Blanke 1985: 11). Curiously enough, Janton in his classification speaks about simplified ethnic languages. This is what is understood as controlled languages on the list of artificial constructs. Klaus Schubert (2011: 52n.) points out that those might be divided into two groups: one from before 1960, that is ethnic languages strongly simplified for the sake of international communication, such as Peano s Latino sine flexione and Ogden s Basic English, and another from after 1960, which comprises languages created to facilitate industrial communication, e.g. Caterpillar Fundamental English or ScaniaSwedish (see Schubert 2001). This raises the question of why otherwise natural languages would then be described as artificial. In accordance with the previous definition, these languages result
128 Ida Stria from a process of deliberate and conscious planning; furthermore, they are designed by a particular individual or an identifiable group of individuals. By the same token, this understanding of the term artificial languages includes several natural types, such as trans-ethnic languages (those that have lost their ethnic ties by becoming internationally spread, such as Medieval Latin), standard forms (written German), revitalized languages (Modern Hebrew, Hawaiian) as well as pidgins and creoles. It might therefore be concluded that there is no binary opposition of natural vs. artificial, but instead there is a continuum of deliberate influence as presented in Schubert 1989: 22 or Koutny 2009: 118, that is, from an artificial a priori system to an ideal untouched ethnic languages (fig. 1). Ideal natural language Frisian Standard German Korean/Hungarian Nynorsk/Indonesian Modern Hebrew/Cornish Occidental/Interlingua standard literary languages small changes (often in one aspect only) extensive changes in many aspects; far-reaching planning revitalization ( reinvention ) naturalistic a posteriori Esperanto schematic a posteriori Volapük mixed system based on natural languages Loglan Solresol programming languages mixed system statistically derived from natural languages a priori Artificial language Figure 1 Scale of artificiality. Based on Schubert 1989: 22 and Koutny 2009: 118. In the case of modern standard varieties, as Duličenko (1989: 53) puts it, a continuous a posteriorization of heterodialectal, but nevertheless monolingual material can be found (see also Schubert 1989: 9). What is more, revitalized languages undergo such
Classifications of artificial languages 129 vast changes that the historical continuity often assumed as required to deem a language as natural is broken and the older as well as the newer form of the language might be considered as two separate although related languages as it is in the case of Biblical and Modern Hebrew (Romaine 2011: 186n.). Thus Romaine (2011) proposes a new label for such languages: reinvented. It is worth noting that most of these so-called natural languages do not in fact have any native speakers. Rural Breton speakers use a completely different version of the language than do speakers of Neo-Breton, who have learnt it as a second, or third language (Romaine 2011: 217n.). 2. The Curious Case of Pidgins Among the languages that stand somewhere between the poles of artificiality and naturalness are pidgins and creoles collectively called contact languages. To decide where exactly they should be put on the scale (fig. 1) the characteristics of a natural language should first be listed. On the basis of Duličenko 1989 and Liu 2001 an ethnic language can be described as a basic conventionalized system that expands with time to new domains, is linked to a certain mostly monolingual population in a given area and requires a constant interaction between speakers to be passed on to next generations. Out of these criteria of naturalness pidgins, creoles and planned languages fulfill only a few, and each type fulfills a different number. requirement ethnic pidgin creole planned development to yes no (unless it yes yes if socialized new domains changes into an expanded pidgin) linked to a specific yes yes, to several yes no population linked to a territory yes yes yes no first language and monolingual native speakers planning designed for international communication yes no yes no (Esperanto has native speakers, but none are monolingual) no (yes in later stages of development) no (may be used as such) no yes no (possibly in later stages of development) no (may be used as such) first form spoken spoken spoken written yes (not so strict in later stages of development) yes
130 Ida Stria As can be seen, pidgins share the most with planned languages, although in terms of artificiality they might be regarded as pure because their coming into existence is through an unconscious, linguistically naive process with no particular creator. The latter can also be said about creoles. Therefore in fig. 1 contact languages should be placed separately, i.e. creoles in the same place as natural languages (because the only difference between the two is their source) and pidgins perhaps after languages where conscious planning has been involved (Nynorsk) and those which have been revitalized (where changes have been introduced on the basis of a language already in existence, for example Cornish). 3. Functional classification The basic functional classification of artificial languages can be again summed up in a binary division into quasi-langue, i.e. (fictional) projects as well as parole, i.e. full-fledged languages (Blanke 2001: 52). This categorization brings about a further classification according to the purpose of their creation: IALs, artistic languages (artlangs) created for aesthetic reasons, constructed languages (conlangs) invented to exercise the limits of language, experimental languages to exercise a philosophical idea. Of course, a language classified in one place might be easily moved elsewhere. For example Toki Pona usually said to be an experimental construct created to help express positive ideas might indeed be considered as an IAL because its first goal was that people communicate. Láadan was firstly created as a part of a fictional world in a book, where it was intended to be used initially by women to express the perceptions of women not lexicalized in existing languages. Unfortunately conlangs and artlangs are not treated very seriously by linguists they are seen rather as games than real languages. Among these groups Klingon and Tolkien s creations have been nevertheless described from a somewhat linguistic perspective in Okrent 2010 and Adams 2011 (Klingon is also the subject of two theses available on the Internet: Hermans 1999 and Wahlgren 2004). Why they seem so interesting is that they grow in popularity with an ever expanding group of supporters. Therefore the birth of a Klingon communicative community might be taking place. Such a community, as shown in section 2, is a requirement to change a mere project into a functioning language. It is also step 12 in an actual functional classification of Blanke (2001: 53-57). He described 28 levels (the first version had only 19) through which a language (project?) must go to attain the final stage of development. Among these the existence of journals, original literature and specialized vocabulary can be found. According to these criteria three types of planned languages can be distinguished: projects (Plansprachenprojekte),
Classifications of artificial languages 131 planned semi-languages (Semiplansprachen), planned languages (Plansprachen). Blanke applied his criteria only to planned languages (IALs), but this list can also be used with regard to other languages. If one of them went through, for example, 14 of the stages, it might be considered a semi-language. One of the last steps to becoming a real language would be the appearance of native speakers. This categorization based on usage is sociolinguistically useful, although it must be remembered that it can only be applied to languages designed to facilitate international communication. As Liu (2001: 131) rightly points out [...] the goals of creoles and planned languages are not the same, so if we evaluate pidgin development according to Blanke s criteria, a creole language would not be a developed language. The same is true for quite a great number of ethnic languages. What is more, even Esperanto thought to be a fully realized language with native speakers does not meet the previous requirement of monolingual speakers interacting with each other. The so-called denaskaj esperantistoj (who are always at least bilingual) do not interact with each other on a regular basis and have no actual influence on the language. This means that Esperanto cannot be considered as creolized. Interestingly though, this situation is in fact regarded as an advantage for learners as they can become fully accepted language community members and those who master the language will be treated with high esteem (see Schubert 1989: 13). 4. Conclusions The present paper deals with some of the problems with respect to a plethora of types of artificial languages as well as a smaller number of diverse classifications of them. As has been shown, the term artificial pertains to constructs ranging from such restricted closed codes as machine languages to several entities widely regarded as natural. Here the name is used to describe every system that has been created and planned consciously and in order to be used in human-to-human communication. As it turns out, such an understanding has never produced any other classification than that of Couturat and Leau (in fact, it only related to IALs; it is the author s belief that it can be successfully used to categorize all of the languages mentioned). The functional classification developed by Blanke may be applied to many languages, but it can only validly assess planned languages with broader goals than some of the other artificial languages. Therefore in accordance with some interlinguists an extended scale of artificiality is proposed. Research on artificial languages should be expanded to all those languages that have a functioning community. Of course, most research would be still conducted in relation to planned languages, but it is important to not overlook the moment other such communities are formed (e.g. Lojban, Klingon).
132 Ida Stria Bibliography Adams, Michael (ed.) 2011: From Elvish to Klingon: Exploring Invented Languages. Oxford, New York: Oxford University Press. Blanke, Detlev 1985: Internationale Plansprachen. Eine Einführung. Berlin: Akademie Verlag. Blanke, Detlev 1989: Planned languages a survey of some of the main problems. In: Schubert, K. (ed.): Interlinguistics: Aspects of the Science of Planned Languages. Berlin, New York: Walter de Gruyter. 63 87. Blanke, Detlev 1997: The Term Planned Language. In: Tonkin, H. (ed.): Esperanto, Interlinguistics, and Planned Language. University Press of America. 1 20. Blanke, Detlev 2001: Vom Entwurf zur Sprache. In: Schubert, K. (ed.): Planned Languages: From Concept to Reality. Interface. Journal of Applied Linguistics 15.1. 37-89. Carlevaro, Tazio 1989: Planned auxiliary language and communicative competence. In: Schubert, K. (ed.): Interlinguistics: Aspects of the Science of Planned Languages. Berlin, New York: Walter de Gruyter. 173 187. Duličenko, Aleksandr 1989: Ethnic languages and planned language. In: Schubert, K. (ed.): Interlinguistics: Aspects of the Science of Planned Languages. Berlin, New York: Walter de Gruyter. 47 61. Eco, Umberto 1997: The search for the perfect language. London: Fontana Press. Hermans, Judith 1999: Klingon and its Users: a sociolinguistic profile. Tilburg University. «http:// www.judion.de/klingon/». Koutny, Ilona 2009: Esperanto im Rahmen der Sprachtypologie. In: Fiedler S. (ed.): Esperanto und andere Sprachen im Vergleich. Beiträge der 18. Jahrestagung der Gesellschaft für Interlinguistik e.v., 21. 23. November 2008, in Berlin. 117 130. Liu, Haitao 2001: Creoles, Pidgins, and Planned Languages. In: Schubert, K. (ed.): Planned Languages: From Concept to Reality. Interface. Journal of Applied Linguistics 15.1. 121 177. Okrent, Arika 2010: In the Land of Invented Languages: Esperanto Rock Stars, Klingon Poets, Loglan Lovers, and the Mad Dreamers Who Tried to Build A Perfect Language. New York: Spiegel & Grau. Romaine, Suzanne 2011: Revitalized Languages as Invented Languages. In: Adams, M. (ed.): From Elvish to Klingon: Exploring Invented Languages. Oxford, New York: Oxford University Press. Schubert, Klaus 1989: Interlinguistics its aims, its achievements, and its place in language science. In: Schubert, K. (ed.): Interlinguistics: Aspects of the Science of Planned Languages. Berlin, New York: Walter de Gruyter. 7 44. Schubert, Klaus 2001: Gestaltete Sprache. Plansprachen und die regulierten Sprachen der internationalen Fachkommunikation. In: Schubert, K. (ed.): Planned Languages: From Concept to Reality. Interface. Journal of Applied Linguistics 15.1. 223 257. Schubert, Klaus 2011: Zum bewussten Eingreifen in die Sprache. In: Brosch, C. & S. Fiedler (eds.): Florilegium Interlinguisticum. Festschrift für Detlev Blanke zum 70. Geburtstag. Frankfurt am Main, Berlin, Bern, Bruxelles, New York, Oxford, Wien: Peter Lang. 47 60. Wahlgren, Yens 2004: Klingon as Linguistic Capital. A Sociologic Study of Nineteen Advanced Klingonists. Lund University. «http://www.angelfire.com/trek/yensw/pdf/thesis.pdf».